Recording Industry, Technology of
RECORDING INDUSTRY, TECHNOLOGY OF
Thomas Edison never envisioned that the phonograph that he invented would ever be used for entertainment. He saw it as a business dictation machine or as a telephone answering machine, not as a music player. However, when he made the first acoustic music recordings on wax cylinders in 1877, Edison unwittingly unleashed the entertainment potential of prerecorded music. The possibility of a library of music recordings being available for personal use sparked a frenzy of recording that has continued unabated.
Many of the early electrical recording studios were essentially radio station facilities. The equipment that was so vital to the broadcasting industry was quickly adopted by the entertainment industry to record music as well as soundtracks for motion pictures. Whether broadcasting or making a record, the sequence of events was the same—performers and musicians created sounds that were gathered by microphones and transformed into electrical signals. These signals were sent to an audio mixer, where they were amplified and combined. Engineers monitored the signals, adjusted their levels, and mixed them into a single-channel output. This output was either recorded on a storage medium or broadcast live. An acetate disc was the preferred recording medium until magnetic tape was developed in the late 1940s.
The technical evolution of the music recording studio was slow, however, compared to the film industry. When Walt Disney's film Fantasia premiered in 1940, the soundtrack had been recorded on nine separate channels that were mixed down to four channels for theatrical presentation. Multiple-channel recording in the audio studio, however, would not be adopted until stereophonic (two-channel) sound was introduced to the public in the 1950s. Regardless of the format, the components in any recording studio signal chain were, and still are, microphones, audio consoles, recording equipment, and monitor loudspeakers. Peripheral equipment used to modify the audio signal includes equalizers, compressors, limiters, and reverberation chambers.
Microphones and Loudspeakers
Mechanical transducers, such as microphones and loudspeakers, are often considered to be the most critical links in the recording chain. They must accurately convert acoustical sound waves into electrical signals and vice versa. When sound waves strike the diaphragm of a microphone, small electrical currents are created that are proportional to the amount of diaphragm movement. This current, or signal, must be amplified for use. There are literally dozens of microphone types with varying electrical and physical characteristics, ranging from a rather large studio condenser microphone to a subminiature lavalier microphone that is the size of a pencil eraser.
Several characteristics that describe microphones include the type of diaphragm element, the directional pattern of sensitivity, and the frequency response. The element types for most professional applications are dynamic, ribbon, and condenser. Dynamic microphones come in all sizes and shapes, and they are very rugged and utilitarian. Ribbon microphones are more fragile, are more sensitive to low-level sounds, and have a better frequency response than do dynamic microphones. Condenser microphones generally have the widest frequency response, and they are preferred for studio use. They have built-in amplifiers that require external or internal power supplies to operate.
Microphones are sensitive to sounds coming from different directions. Directional sensitivity patterns in the shape of a heart (one direction), a figure-eight (two directions), and a circle (all directions) correspond to the pattern names of unidirectional, bidirectional, and omnidirectional, respectively. Some microphones have variable patterns, and a switch can be used to change the directional characteristics.
The human hearing range or audio spectrum is generally considered to go from 20 Hz (hertz, a unit of frequency) to 20,000 Hz. Although the best microphones should have a comparable frequency response, many sonic variations are acceptable and may even be desirable. Microphones with increased treble response can add brightness in some applications, while those with an increased bass response can enhance vocal performances.
Monitor loudspeakers should be able to reproduce the same audio spectrum but at volumes approaching rock concert level. Sonic variations are not desirable; they can actually overemphasize or mask portions of the musical material, making accurate audio evaluations in the control room somewhat difficult. Many speaker enclosures contain three or more loudspeakers called tweeters, midranges, and woofers. Each is designed to reproduce only a portion of the spectrum: high, middle, or low frequencies, respectively. Some loudspeaker systems use separate amplifiers for high and low frequencies, as well as multiple speakers for each frequency range to ensure the most accurate sound reproduction possible.
The Audio Console
The complexity of an audio recording console may initially be overwhelming, but its basic operations are relatively simple. Regardless of the console size or layout design, there are provisions for assigning inputs to outputs, adjusting the input and output levels, watching the levels on meters, and listening to the entire process on control room loudspeakers. A typical 1960s audio production console contained 24 input modules— each identical to the other—feeding 16 outputs. There were literally hundreds of knobs, switches, meters, and dials spread all over the seven-foot-long control board. Contemporary consoles can contain as many as 96 input modules feeding 48 outputs.
During recording, the signal from a microphone goes to an input/output (I/O) module or channel strip, and its volume and equalization are adjusted. The signal is then assigned to an output bus, or a pan control sends the signal to two output buses to create a stereo effect. Each I/O module includes auxiliary outputs (sends) either for external effects such as echo or for monitoring requirements such as headphone mixes for performers. A special patch point on each I/O module, called an insert jack, permits connection of external signal-processing devices. Compressors or limiters can control input signal dynamics by reducing sudden, extremely loud sounds to more manageable audio levels. Effects processors can create reverberation or delays, or modify virtually any aspect of an audio signal, while different types of equalizers increase or decrease portions of the audio signal. Stanley Alten (1999) separates processors into four categories: spectrum processors that affect the overall tonal balance, time processors that affect the time interval between a signal and its repetition, amplitude processors that affect a signal's dynamic range, and noise processors that reduce tape noise. A VU (volume unit) meter monitors the signal level of each input module, as well as the output buses. Each output bus is connected to a multitrack audio recorder.
After recording, the multitrack master is played back through the same console so the engineer can mix down or combine all the channels into a new master recording. The same signal processing that is available during recording can be used during the mixdown process as well. Trying to get the precise balance, equalization, effects, and stereo panning when mixing down 24 or 48 channels requires hundreds of level and control changes during a short recording. Prior to the development of console automation, recording engineers typically kept detailed logs of all the settings, and numerous rehearsals were required to get the right mix. With automation, a computer captures and stores each console setting change automatically, which greatly simplifies the mixdown process.
The Magnetic Tape Recorder
When Elvis Presley recorded his first demonstration tapes in 1954, an old six-input radio mixer and two single-track recorders were all that was needed to launch the career of the "King of Rock-n-Roll." The Beatles recorded their first songs on a two-track tape recorder in 1962, putting all of the instrumentals on one track and the vocals on the second track. As artists demanded more creative flexibility, 16-and 24-track audio recorders using two-inch tape were developed, and by the 1970s, 32-track recorders were common.
Regardless of the number of tracks, all magnetic tape recorders operate on the same principles. A spool of iron-particle recording tape passes across three heads that are aligned in a tape transport. The tape is drawn at a precise speed by the drive mechanism. Each head contains a coil of wire wrapped around layers of steel, called poles. A very narrow gap between the poles focuses a magnetic field on a portion of the recording tape. The first head erases a track on the tape, the second head records a new audio signal on the track that has just been erased, and the third head plays back the recorded track.
A typical professional magnetic recorder can handle 70 dB (decibels, a unit of loudness) of dynamic range. Because the dynamic range of some music—measured as the difference between the loudest passages and no sound—can exceed 120 dB, the quietest passages can be lost in electronic noise. Each time a tape is played back and copied or re-recorded, the level of noise increases relative to the signal. The first noise reduction system, named Dolby after its inventor (Ray Dolby), improved a recorder's signal-to-noise ratio by as much as 20 dB, permitting more extensive re-recording or overdubbing.
Analog audio signals consist of continuously changing information with two characteristics: duration (or time) and volume (or amplitude). For example, a musical score visually represents notes of various frequencies that are played over time. The performer determines the appropriate amplitude. This musical information must be converted to digital data for processing and storage. The musical characteristics of time and amplitude correspond to the digital audio characteristics of sampling and quantization in the analog-to-digital (A/D) conversion process.
Sampling takes a snapshot of a musical moment in time, while quantization assigns numerical values to the snapshot information. The sampling rate must be able to capture successfully the highest frequency in the audio spectrum while avoiding sampling errors, so sampling rates have been standardized at 32, 44.1, and 48 kHz (kilohertz) for different digital audio applications.
Quantization uses binary numbers—ones and zeros—to measure the amplitude of an audio signal. These numbers are formed into words, or bits. The bit length determines the accuracy of a measurement. It actually takes a 16-bit word of 65,536 steps to quantize most audio signals. More accurate quantization calls for 20-or 24-bit words (more than 16 million steps). Reverberation devices, equalizers, and dynamics processors often use a 32-bit system to ensure the highest fidelity audio processing.
Along with multiple standards for sampling and quantization to convert analog audio to digital information, there are also standards for communication among digital devices. Signals sent from one machine to another can remain as digital data, rather than having to be converted to analog audio signals for transmission. There are two transmission standards or protocols: professional equipment uses AES/EBU (Audio Engineering Society/European Broadcast Union), while consumer equipment uses S/PDIF (Sony/Philips Digital Interface) for machine-to-machine communication. With one of these protocols, digital signals can be distributed to a number of devices without any degradation or digital-to-analog (D/A) conversion.
Digital audio consoles follow the form and function of analog audio consoles, although the inputs are converted directly into digital data before any channel assignment or signal processing occurs. Many consoles feature built-in signal dynamics and effects processing in each channel, rather than having to patch external devices into the desired channels. Full remote control is possible via the musical instrument digital interface (MIDI) protocol, working with a MIDI controller, sequencer, or computer. The output signals are in the AES/EBU format that can be connected directly to a digital audio recorder, although they must be converted to analog signals for monitoring purposes.
Innovative designs of the latest digital consoles permit I/O module cluster swapping from one console to another, enlarging a 48-channel console to 96 channels or more for a particular recording session. Some digital consoles offer an expansion capability of 200 channels or more. All the controls of any I/O module can be replicated on a central control module, so the recording engineer does not have to move constantly from one end of a ten-foot-long console to the other while making adjustments. Sampling frequencies are as high as 96 kHz to ensure the utmost fidelity.
Some digital audiotape machines are based on analog open-reel tape recorder transports that use fixed audio heads (the digital audio stationary head, or DASH, format). Others are based on videotape transports that use rotating heads and cassette tapes. A third group of digital recorders does not use tape at all; these record directly to computer hard drives or other portable storage media such as ZIP disks or magneto-optical (MO) disks.
The DASH machines appear similar to analog machines, and they retain many analog tape machine functions, including editing electronically or by the traditional cut-and-splice method. Either 24 or 48 tracks can be recorded on half-inch tape at 30 ips (inches per second).
The rotary-digital-audiotape (R-DAT or DAT) system has become one of the more popular formats. The miniature size of the DAT tape—about half the size of a standard audiocassette—belies the capabilities of the recording system. The DAT features two record tracks, digital inputs and outputs, high-speed search and cueing, and the capability to record time code for video production. Portable recorders with many of these features can be as small as a paperback book.
Multitrack digital tape recorders, also called modular digital multitrack (MDM) recorders, use two manufacturers' incompatible standards: ADAT that records on standard S-VHS videotape and DTRS that records on Hi-8-mm videotape. Both types of MDMs feature 8 tracks of recording, external synchronization for videotape editing, and the ability to link several machines together to create a virtual 128-track digital audio recorder (the modular aspect of the name).
Newer machines record directly to a removable computer hard disk, thereby eliminating tape altogether. Some hard-disk recorders feature up to 24 tracks while retaining many of the MDM features. The more advanced hard-disk recorders are capable of 48 tracks, using 24-bit resolution and 96 kHz sampling. Still another type of digital recording format uses various-sized MO disks as the storage media. Personal mixer/recorders use the 2.5-inch Mini Disc to mix and record either 2 tracks or 8 tracks, while the professional MiniDisc recorders have replaced the venerable endless-loop tape cartridges for many broadcast applications. Consumer MiniDisc recorders and players are replacing analog cassette machines as the preferred portable personal format.
The digital audio workstation (DAW) is a computer-based audio recording and editing system that replicates every function of an entire recording studio—from the audio console, equalizers, compressors, and effects units to the multi-track recorders and editing controllers. Some systems are designed as computer software programs with plug-in boards for computers (host-based), while others feature custom control surfaces and function as stand-alone units. Various storage media are used with the DAW. These can be MO, ZIP, or internal hard disks, or they can be external MDMs. Because the DAW is a totally integrated system that includes recording, editing, and signal routing and processing, speed of operation and flexibility are superior to similar analog equipment. Control of external tape machines or video recorders, as well as MIDI control of electronic music devices, is often included.
Because the ultimate goal of any studio is to capture an artist's performance with the greatest fidelity and to provide the technical and creative tools that are necessary to produce the final mix, acquiring the right assemblage of analog hardware and digital software can be a never-ending quest. The tools can be a multimillion-dollar studio complex, an inexpensive, personal portable mixer/recorder, or a computer. When producer George Martin worked with the Beatles to create the Sgt. Pepper's Lonely Hearts Club Band album in 1967, more than seven hundred hours were spent recording, mixing, and editing on 4-track analog tape recorders to finish the project. Using modern digital equipment, Martin could have significantly reduced the amount of time spent producing that classic recording.
See also:Disney, Walt; Edison, Thomas Alva;Film Industry, Technology of; Radio Broadcasting, Technology of; Recording Industry; Recording Industry, History of; Recording Industry, Production Processes of.
Alldrin, Loren. (1997). The Home Studio Guide to Microphones. Emeryville, CA: MixBooks.
Alten, Stanley R. (1999). Audio in Media, 5th edition.Belmont, CA: Wadsworth.
Eargle, John. (1992). Handbook of Recording Engineering, 2nd edition. New York: Van Nostrand Reinhold.
Gelatt, Roland. (1977). The Fabulous Phonograph: 1877-1977, 2nd edition (revised). New York: Macmillan.
Millard, Andre. (1995). America on Record. Cambridge, Eng.: Cambridge University Press.
Pohlman, Ken H. (1998). Principles of Digital Audio,3rd edition. New York: McGraw-Hill.
Read, Oliver, and Welch, Walter L. (1959). From Tinfoil to Stereo. Indianapolis, IN: H. W. Sams.
Runstein, Robert E. (1974). Modern Recording Techniques. Indianapolis, IN: H. W. Sams.
Runstein, Robert E., and Huber, David M. (1997). Modern Recording Techniques, 4th edition. Boston: Focal Press.
Schoenherr, Stephen. (2000). "Recording Technology History." <http://history.acusd.edu/gen/recording/notes.html>.
Woram, John M. (1989). Sound Recording Handbook. Indianapolis, IN: H. W. Sams.
John M. Hoerner, Jr.